Findings

Pathogenic Escherichia coli is the causative agent of colibacillosis that brings severe clinical sighs such as diarrhea, meningitis and sepsis in domestic animals. E. coli can be divided into four phylogenetic groups using multiplex PCR method [1]. Several authors have analyzed the distribution of the main phylogenetic groups among E. coli strains isolated from human and animals. Extraintestinal pathogenic E. coli usually belongs to groups B2 and D, while the commensal strains are groups A and B1 in humans [2]. In healthy food-producing animals, a predominant distribution of group B1 was reported in E. coli from cattle, while group A was predominantly prevalent in pigs and chickens [3–5]. On the contrary, although predominant phylogenetic groups were reported in E. coli from diseased poultry [6–8], little information regarding to diseased cattle and pigs was available.

Antimicrobial agents have been widely used for treatment of colibacillosis in domestic animals. Antimicrobial resistance is more frequently found in E. coli isolates from diseased animals than apparently healthy animals [9]. Increases in cephalosporin-resistant E. coli in domestic animals have been a significant worldwide concern including Japan. Extended-spectrum β-lactamase (ESBL)-producing isolates of E. coli were identified in domestic animals around 2000 in Japan [10, 11]. The third and greater generation cephalosporin antibiotics are clinically important antimicrobial agents in human and animal medicines. Such agents are approved to treat bacterial diseases in cattle and pigs and not licensed for enhancing animal growth in Japan. However, cephalosporin-resistant E. coli has been identified in isolates from not only cattle and pigs but also poultry in Japan [10, 12]. At present the reason for the prevalence of cephalosporin-resistant E. coli in poultry is unclear.

In the present study, we determined the distribution of phylogenetic groups in E. coli from food-producing animals affected with colibacillosis in Japan. In addition, we examined the characteristics of cephalosporin-resistant isolates of E. coli from the diseased animals.

A total of 318 E. coli isolates from different animals affected with colibacillosis between 2001 and 2006 throughout Japan were used in this study: 72 bovine isolates, 89 poultry isolates and 157 porcine isolates in 23, 20 and 28 prefectures, respectively, located from north to south Japan. All the strains were stored in 10% skimmed milk at -80°C until use. The phylogenetic grouping was performed by multiplex PCR as described by Clermont et al. [1].

The minimum inhibitory concentration (MIC) of cefazolin (CEZ) was determined by an agar dilution method in accordance with the guidelines of the Clinical Laboratory Standards Institute (CLSI) [13, 14]. The MIC of some of the isolates to CEZ was previously determined [15]. Susceptibilities of CEZ-resistant isolates to several β-lactamase antibiotics, cefpodoxime, cefotaxime, ceftazidime, aztreonam, imipenem and meropenem were tested according to the CLSI guideline [13, 14] using the commercially available broth microdilution test (Eiken Co. LTD, Tokyo, Japan). E. coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were used as quality controls for MIC determination. The Chi-square test was performed on the antimicrobial susceptibility and the phylogenetic subgroup data. The resistance rates in each phylogenetic group were compared with those in the overall strains tested. For each comparison, a P value of < 0.05 was considered to denote significant differences.

A double-disk synergy test for detection of ESBLs was performed using clavulanate and cefotaxime, ceftazidime, cefpodoxime, or aztreonam disks (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) as previously described [10]. Detection of β-lactamase genes was carried out by PCR using primers for bla TEM , bla SHV and bla PSE-1 as described previously [10]. Multiplex PCR was applied to the AmpC genes, groups ACC, FOX, MOX, CIT and EBC, detection as previously described [16]. As for samples positive for CIT group, sequence analysis was performed using the primer pairs for amplification of the full-length gene of bla CMY-2 [10]. Multiplex PCR was used for bla CTX-M detection as previously described [17]. Subtypes of CTX-M β-lactamases were determined using the primer pairs as follows: CTX-MF (5'-GACTATTCATGTTGTTGTTATTTC-3') and CTX-MR (5'-TTACAAACCGTTGGTGACG-3') for CTX-M-group1 [18]; bla CTX-M-2 F (5'-ATGATGACTCAGAGCATTCG-3') and bla CTX-M-2 R (5'-TCAGAAACCGTGGGTTACGA-3') for CTX-M-group2 [10]; CTXM825F (5'-CGCTTTGCCATGTGCAGCACC-3') [19] and CTXM20 (5'-ATAACCGTCGGTGACAATT-3') [17] for CTX-M-group8 and 25/26. Nucleotide sequences were preformed directly on both strands of PCR products by using dye terminator chemistry. The DNA alignments and deduced amino acid sequences were examined using the BLAST program (National Center for Biotechnology Information, USA).

The phylogenetic analysis revealed that 318 isolates of E. coli from diseased animals were classified into four phylogenetic groups (Table 1). The phylogenetic group A was commonly predominant in isolates from cattle (36/72, 50%) and pigs (101/157, 64.3%) whereas groups A (44/89, 49.4%) and D (40/89, 44.9%) were predominant in isolates from poultry. The B2 group was rarely found in E. coli from healthy cattle, chickens and pigs in Brazil [3] as well as Japan. In Korea, group B2 was also not found in E. coli isolates from food-producing animals [4]. These trends were similar to those of the present study although clinical isolates from food-producing animals were used here. On the other hand, in the United States, groups A and D were predominant (38% and 28%, respectively) in E. coli from diseased poultry and group B2 was also found in 19% [6]. In European countries, group B2 was often isolated from diseased poultry [7, 8]. Thus, the distribution of phylogenetic groups may be determined not only by the animal species but also by their health status or geographical region.

Table 1 Distribution of Escherichia coli isolate phylogenetic groups from diseased animals in Japan between 2001 and 2006
Full size table

Previous studies showed that the phylogenetic group distribution may be related to antimicrobial resistance prevalence in human isolates of E. coli [20]. In the present study, 38 (11.9%) isolates exhibited CEZ resistance (MIC: ≥32 μg/ml) (Table 2). Twenty-six (68.4%) of the CEZ-resistant isolates belonged to group A. However, rates of CEZ resistance among the phylogenetic groups, except for group B2, were in 8.6-14.4%. Thus, no significant difference in the resistance rates between the phylogenetic groups was observed.

Table 2 MICs of β-lactam antibiotics for 38 cefazolin-resistant isolates of E. coli from diseased animals isolated between 2002 and 2006
Full size table

Of the 38 resistant isolates, 11 isolates from two cattle and nine poultry produced ESBLs while the remaining 27 isolates produced AmpC β-lactamase. The two bovine isolates expressed bla CTX-M-2 , while, of the nine poultry, four expressed bla CTX-M-25 , three expressed bla SHV-2 , one expressed bla CTX-M-15 and one expressed bla CTX-M-2 . The bla TEM-1 gene, being non-ESBL, was detected in 12 isolates. This is the first description of the bla SHV-2 , bla CTX-M-15 and bla CTX-M-25 enzymes in E. coli from food-producing animals in Japan. In human patients in Japan, E. coli strains carrying bla CTX-M-2 were predominantly prevalent around 1997 [21], and then the dominant β-lactamase type has changed to the CTX-M-9 group, including bla CTX-M-14 and bla CTX-M-16 , since 2001 [22]. As for food-producing animals, E. coli carrying bla CTX-M-2 have emerged in apparently healthy broilers [10] and cattle [11] in 2000 and pigs in 2003 [12]. In 2002, bla CTX-M-18 , identical to amino acid sequence of bla CTX-M-14 , was identified in cephalosporin-resistant isolates from healthy broilers [10]. On the other hand, of the remaining 27 AmpC β-lactamase producing isolates, 13 isolates were PCR positive for CIT group, being identified bla CMY-2 : 10 from poultry, two from cattle and one from swine (Table 2). There were no positive signals for group ACC, FOX, MOX and EBC. The bla CMY-2 gene was found to be prevalent in a wide range of food-producing animal species in Japan [10, 12], as well as other countries [23].

In the present study, 5 of 8 isolates producing CTX-M ESBL and all three isolates producing SHV-type ESBL belonged to group A. However, group B2 is predominant in SHV-type ESBL producing E. coli isolates of human origin whereas group D is predominant in CTX-M enzyme-producing isolates [24]. On the contrary, 7 of 13 isolates producing CMY-2 β-lactamase belonged to group A. Group B1 was dominant in CMY enzyme-producing E. coli from retail broiler meats [25]. It is likely that the genotype of E. coli carrying β-lactamase genes may be different between humans and food-producing animals.

It is essential to understand the prevalence of resistance to clinically important antimicrobials in bacteria from food-producing animals. The MICs of six β-lactam antibiotics are summarized in Table 2. All CEZ-resistant isolates were susceptible to imipenem and meropenem. All isolates exhibit cefpodoxime resistance (MIC of breakpoint: ≥8 μg/ml) except for one porcine isolate in which β-lactamase expression was not determined. All isolates expressing CTX-M enzymes exhibit cefotaxime resistance (MIC of breakpoint: ≥64 μg/ml) with the exception of one poultry isolate (MIC: 16 μg/ml) harboring bla CTX-M-25 . Resistance to ceftazidime was found in three poultry and one porcine isolate with bla CMY-2 , and one poultry isolate with bla CTX-M-15 . Resistance to aztreonam was found in a single bovine isolate with bla CTX-M-2 . The MICs of antimicrobials tested varied in isolates with each β-lactamase gene.

In conclusion, our study suggests that the distribution of phylogenetic groups in E. coli from diseased domestic animals varies regionally in addition to animal species. Moreover, ESBL-producing E. coli from the animals were frequently assigned to a specific group, being different from human ESBL isolates. Further study on ESBL-producing E. coli from several animal sources using the PCR-based phylogenetic typing and the other genotyping methods may clarify the relationship of the resistant bacteria and/or the resistance determinants prevalent in the food-chain and the environments.